How Predictive Modeling can save your crops from disease
By Brook Sauer Ruiz, founder and CEO of Elby’s Organics,
a new food startup based in Hamburg, Germany.
What is predictive modeling?
Predictive modeling is exactly what the name implies. It’s the method of using big data to create models that predict the health and volume of crop yield.
In short, predictive analytics uses business intelligence to first collect, integrate, and analyze big farming data. From there, models are produced that predict conditions pests are most likely to strike. This allows farmers to make precise decisions about where and when crops need pesticides.
Why is it important?
At its core, the method of Predictive Modeling helps in decision making. In other words, it’s a “crystal ball” for farmers to see into the future and anticipate potential problems. They then can use this information to make informed decisions that ultimately save time and money.
Indeed, pest infestations can be devastating to a crop yield. Traditionally, preventing infestations meant doing a blanket sweep of pesticides across entire fields of crops. However, this approach uses vast amounts of pesticides to control a problem that hasn’t happened yet. Moreover, the time needed to spray, and the use of machinery comes at a cost that can be reduced with the help of predictive modeling.
Examples of Predictive Modeling in Europe
In Spain, this type of modeling is gaining attention at a national level. In 2017, The Andalusian Plant Protection and Information Network (RAIF) even participated in a bold program to benefit the Integrated Olive Production Association (APIs).
Simply put, big data gathered by the RAIF is analyzed and formulated into a prediction on what percentage of olives would be eaten by flies. The API member farms (1,568 members) then receive these weekly predictions and translate them into an informed pest management strategy.
The University of Kiel has also made bold advances in predictive modeling for wheat, and scientists are currently working on a model for corn.
Video: Take a look at predictive modeling in action in Germany.
The challenges of predictive modeling
First of all, a large amount of data is needed in order to generate predictive models. Secondly, areas of the world that don’t have the infrastructure to support IoT and smart farming solutions are not yet capable of harnessing the full power of this technology.
Along with this, extensive research is needed when looking at how to best prevent pest infestations. Progress can be slow and expensive depending on the research needed.
Partnership programs such as the fly initiative in Spain, are an example of action at the government level that serves to help farmers. This type of top-down support, along with widespread adoption of smart farming solutions drive the cost down and allow for more types of pest control modeling analytics to be possible.
Predictive Modeling shapes the future of farming
Advances in predictive modeling ultimately give farmers the tools they need to make well informed and more sustainable decisions about their resources.
Using predictive models to improve the health and volume of crop yields, should be celebrated and embraced at a local and national level. Data driven decisions across all aspects of farming are the key to positive growth and sustainability in the industry.
We are a creative team that loves writing and sharing its broad range of expertise with you. For our blog, we regularly write articles about current topics in the fields of IT, trade and agriculture. Whether it’s cloud solutions, artificial intelligence, digitalization of business processes or smart farming – we keep our finger on the pulse and you up to date: with exciting reportages, interesting interviews, in-depth background reports and illustrative overviews.